Unique Somatodendritic Distribution Pattern of Kv4.2 Channels On

View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Repository of the Academy's Library

European Journal of Neuroscience

European Journal of Neuroscience, Vol. 35, pp. 66–75, 2012 doi:10.1111/j.1460-9568.2011.07907.x

SYNAPTIC MECHANISMS

Unique somato-dendritic distribution pattern of Kv4.2 channels on hippocampal CA1 pyramidal cells

Katalin Kerti, Andrea Lorincz and Zoltan Nusser Laboratory of Cellular Neurophysiology, Institute of Experimental Medicine, Hungarian Academy of Sciences, Budapest, Hungary

Keywords: dendrite, hippocampus, immunohistochemistry, potassium channels, pyramidal cell, rat

Abstract + A-type K current (IA) plays a critical role in controlling the excitability of pyramidal cell (PC) dendrites. In vitro dendritic patch-pipette recordings have demonstrated a prominent, sixfold increase in IA density along the main apical dendrites of rat hippocampal CA1 PCs. In these cells, IA is mediated by Kv4.2 subunits, whose precise subcellular distribution and densities in small-diameter oblique dendrites and dendritic spines are still unknown. Here we examined the densities of the Kv4.2 subunit in 13 axo-somato-dendritic compartments of CA1 PCs using a highly sensitive, high-resolution quantitative immunogold localization method (sodium dodecyl sulphate-digested freeze-fracture replica-labelling). Only an approximately 70% increase in Kv4.2 immunogold density was observed along the proximo-distal axis of main apical dendrites in the stratum radiatum with a slight decrease in density in stratum lacunosum- moleculare. A similar pattern was detected for all dendritic compartments, including main apical dendrites, small-diameter oblique dendrites and dendritic spines. The specificity of the somato-dendritic labelling was confirmed in Kv4.2) ⁄ ) tissue. No specific immunolabelling for the Kv4.2 subunit was found in SNAP-25-containing presynaptic axons. Our results demonstrate a novel distribution pattern of a voltage-gated ion channel along the somato-dendritic surface of CA1 PCs, and suggest that the increase in the IA along the proximo-distal axis of PC dendrites cannot be solely explained by a corresponding increase in Kv4.2 channel number.

Introduction Understanding the way nerve cells integrate their synaptic inputs patch-clamp recordings have provided evidence for its presence in requires the knowledge of the precise subcellular location and CA1 PC dendrites, and revealed a sixfold increase in its density as a densities of voltage-gated ion channels (e.g. Na+,Ca2+,K+, mixed function of distance from proximal to distal main apical dendrites cation) on the axo-somato-dendritic surface of the cells. Among the (Hoffman et al., 1997). voltage-gated ion channels, K+ channels (Kv) are molecularly the Using Kv4.2) ⁄ ) mice, Chen et al. (2006) have demonstrated that most diverse and play a powerful role in controlling neuronal in hippocampal CA1 PCs dendritic IA is exclusively mediated by the excitability (Hille, 2001; Gutman et al., 2003). Particular attention Kv4.2 subunit. This is consistent with the results of light microscopic + has been paid to the rapidly inactivating or A-type K current (IA) immunohistochemical studies revealing strong Kv4.2 immunosignal due to their widespread distribution in the CNS (Serodio & Rudy, in the dendritic layers of the CA1 area (Maletic-Savatic et al., 1995; 1998) and their low activation threshold that is around the resting Varga et al., 2000; Rhodes et al., 2004; Jinno et al., 2005). membrane potential of most neurons (Birnbaum et al., 2004). It has However, it has been noticed that the intensity of the immunoper- been elegantly demonstrated that IA has a major role in a large oxidase or ﬂuorescent signal is rather uniform across the stratum variety of dendritic processes, including the control of local Na+ radiatum (SR), with a slight decrease in the stratum lacunosum- spike initiation and propagation (Losonczy et al., 2008), the back- moleculare (SLM). One possible explanation for the discrepancy propagation of axonally generated action potentials into the dendrites between the density of IA and that of Kv4.2 immunolabelling is that (Hoffman et al., 1997; Migliore et al., 1999), synaptic integration the oblique dendrites or dendritic spines or axon terminals contain a and plasticity (Margulis & Tang, 1998; Cash & Yuste, 1999; high density of Kv4.2 subunit, which are superimposed and therefore Watanabe et al., 2002; Frick et al., 2004; Kim & Hoffman, 2008; mask the increasing, but lower, density of Kv4.2 subunit in the main Makara et al., 2009). To fulﬁl all these functional roles, IA must be apical dendrites. present in the dendrites of pyramidal cells (PCs). Direct dendritic Here we aimed to determine whether the known increase in IA density in main apical dendrites of CA1 PC is mirrored by a corresponding increase in the density of Kv4.2 subunit, and to Correspondence: Dr Zoltan Nusser and Dr Andrea Lorincz, as above. quantitatively compare the density of the Kv4.2 subunit in large- and E-mails: [email protected] and [email protected] small-diameter dendritic and axonal compartments. To achieve our aims, we carried out immunogold localization of the Kv4.2 subunit Submitted through the Neuroscience Peer Review Consortium (NPRC) framework. with a highly sensitive, quantitative, electron microscopic immuno- Received 13 September 2011, revised 14 September 2011, accepted 15 September 2011 gold localization technique [sodium dodecyl sulphate-digested freeze-

ª 2011 The Authors. European Journal of Neuroscience ª 2011 Federation of European Neuroscience Societies and Blackwell Publishing Ltd Somato-dendritic distribution of Kv4.2 on CA1 PCs 67 fracture replica-labelling (SDS-FRL); Fujimoto, 1995; Rash et al., gold particles (1 : 100; British Biocell International, Cardiff, UK); or 1998; Masugi-Tokita & Shigemoto, 2007]. goat anti-rabbit IgGs coupled to 10-nm gold particles (1 : 100; British Biocell International). Finally, replicas were rinsed in TBS and distilled water before they were picked up on copper parallel bar grids. Materials and methods Specimens were analysed with a transmission electron microscope All procedures were carried out in accordance with the ethical (JEM-1011, JEOL, Tokyo, Japan) equipped with a high-resolution guidelines of the Institute of Experimental Medicine of the Hungarian bottom-mounted CCD camera (Cantega G2, Olympus Soft Imaging Academy of Sciences, which is in line with the European Union Solution GmbH, Munster, Germany). regulation of animal experimentations.

Quantitative analysis of immunogold labelling for the Kv4.2 Tissue preparation subunit Seven adult male Wistar rats (P25–P41; bred in the Institute of Quantitative analysis of CA1 PC somata and 11 dendritic compart- Experimental Medicine’s Animal Facility), three male wild-type and ments was performed only on complete replicas where the stratum ) ) three Kv4.2 ⁄ mice (129 ⁄ SvEv background; Chen et al., 2006; P68– pyramidale (SP), SR and SLM could be clearly distinguished (n =5 P217; kindly provided by Prof. Daniel Johnston from the University of rats). The subcellular compartments were imaged with a Cantega G2 Texas, Austin, USA) were deeply anaesthetized with ketamine camera at 10 000–12 000 · magnification, and were grouped based (0.5 mL ⁄ 100 g) and then were transcardially perfused with ice-cold on their calculated distance from the SP. The distance of an individual fixative. For immunofluorescent reactions animals were perfused with a process from the SP was calculated from the position of the SP and fixative containing either 2 or 4% paraformaldehyde and 15v ⁄ v% picric from the stage coordinate (X, Y ) of the process using the following acid made up in 0.1 m phosphate buffer (PB) for 20 min. Afterwards, equation: 60 lm coronal forebrain sections were cut with a vibratome (VT1000S; Leica Microsystems, Wetzlar, Germany). For SDS-FRL, animals were PðX ; Y Þd ¼jðy y ÞX þðx x ÞY þ x y perfused with a fixative containing 2% paraformaldehyde and 15v ⁄ v% 2 q1 ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi1 2 2 1 m 2 2 picric acid in 0.1 PB for 16 min. Coronal sections of 80 lm thickness x1 y2j= ðy2 y1Þ þðx1 x2Þ were cut from the dorsal hippocampus with a vibratome and were cryoprotected in 30% glycerol. The position of the SP was determined by an imaginary line connecting two end points of the SP with the coordinates x1, y1 and x2, Immunofluorescent reactions y2. According to this layers were categorized as follows: 0–120 lm: Free-floating sections were blocked in 10% normal goat serum (NGS), proximal SR; 120–240 lm: middle SR; 240–360 lm: distal SR; and followed by an incubation in the solution of rabbit anti-Kv4.2 above 360 lm: SLM. The main apical dendrites, oblique dendrites and (1 : 500; APC-023; Alomone Labs, Jerusalem, Israel) antibody made dendritic spines were grouped according to these criteria. Oblique up in Tris-buffered saline (TBS) containing 2% NGS. Sections were dendrites were identified based on their small diameter and the then incubated in Alexa488- or Cy3-conjugated goat anti-rabbit IgGs presence of at least one emerging spine from the dendritic shaft. made up in TBS containing 2% NGS for 2 h. Images from the CA1 Spines were identified based on either their ultrastructure (e.g. small- region were acquired using a confocal laser-scanning microscope diameter structure emerging from a dendrite; n = 3 rats) or from the (FV1000; Olympus, Tokyo, Japan) with a 20 · objective (UPLAN- presence of a PSD (identified in double-labelling reactions with PSD- SAPO, UIS2, NA = 0.75). 95) on isolated spine heads (n = 2 rats). The density of immunogold particles for the Kv4.2 subunit in dendritic spines was not significantly different (P > 0.05, t-test) between these two groups, therefore they were pooled together. The numbers of analysed profiles and counted SDS-FRL gold particles are provided in Tables 1–3. The GABAA receptor b3 Small blocks from the CA1 region were frozen in a high-pressure subunit was used to identify GABAergic synapses on PC somata and freezing machine (HPM 100, Leica Microsystem) and fractured at dendrites. The immunolabelling for the Kv4.2 subunit was also )135 C in a freeze-fracture machine (BAF060, Leica Microsystem), quantified on axon terminals identified by a high density of gold as described previously (Masugi-Tokita et al., 2007; Lorincz & particles labelling the SNAP-25 in three rats. Statistical comparisons tatistica 8 Nusser, 2010). The fractured tissue surfaces were coated with thin were performed with the S Software (Scientific Comput- layers of carbon (5 nm), platinum (2 nm) and carbon (20 nm). Tissue ing, Rockaway, NJ, USA). debris from the replicas was digested in a solution containing 2.5% SDS and 20% sucrose in TBS (pH 8.3) at 80 C overnight. Following several washes in TBS containing 0.05% bovine serum albumin Control experiments (BSA), replicas were blocked in TBS containing 0.1% BSA for 1 h, Double-labelling experiments were included in the manuscript only if then incubated overnight in blocking solution containing the following the measured background labelling on E-face and the measured primary antibodies: rabbit anti-Kv4.2 (1 : 5000; Alomone Labs); specific P-face labelling for the Kv4.2 subunit on dendritic shafts was mouse anti-postsynaptic density (PSD)-95 (1 : 3000; MAB1598; very similar to that obtained in single-labelling reactions. Furthermore, Chemicon, Billerica, MA, USA); mouse anti-c-aminobutyric acid in the double-labelling experiments the lack of cross-reactivity of (GABA)AR b3 (1 : 800; 75–149; NeuroMab); mouse anti-SNAP-25 secondary antibodies was tested by omitting the primary antibody for (1 : 5000; 111 001; SYSY; Goettingen, Germany). Replicas were then the Kv4.2 subunit and checking the lack of 10-nm gold particles on incubated for 2 h in TBS containing 1% BSA and the following the labelled structures (e.g. PSDs or in axon terminals for PSD-95 or secondary antibodies: goat anti-mouse IgGs coupled to 5-nm or 15-nm SNAP-25, respectively). Lack of labelling demonstrates that the goat

anti-rabbit secondary antibody does not recognize the mouse anti- SNAP-25 or PSD-95 primary antibodies. Non-speciﬁc background

(9, 23) labelling was measured on E-face structures around the measured P- faces as described previously (Lorincz & Nusser, 2010). 20.4 (6, 16) 11.2 To conﬁrm the speciﬁcity of our Kv4.2 immunoreactions on SDS- ) 12.5 (16, 59) digested replicas, the above-described immunogold reactions were 327) 8.8 (7, 15) repeated in three Kv4.2+ ⁄ + and three Kv4.2) ⁄ ) mice, with the

(13, 240) exception that here the layer categorization was as follows: 0–100 lm:

8.9 proximal SR; 100–200 lm: middle SR; 200–300 lm: distal SR; and above 300 lm: SLM. In Kv4.2) ⁄ ) mice the immunogold particle density for the Kv4.2 subunit measured on P-faces was similar to the gold particle density measured on E-faces in replicas obtained from (11, 25) wild-type mice and rats, validating our approach of estimating the

13.6 level of non-speciﬁc labelling on E-faces.

(14, 36) Results Distribution and speciﬁcity of the Kv4.2 subunit 15.1 immunoreactivity in the hippocampal CA1 area Light microscopic immunoﬂuorescent reactions for the Kv4.2 subunit

(10, 19) revealed a rather uniform labelling pattern throughout the stratum oriens and SR of the CA1 area of rat dorsal hippocampus with a slight 11.0 reduction in the intensity in the SLM (Fig. 1A), in line with previous reports (Maletic-Savatic et al., 1995; Varga et al., 2000; Rhodes et al., 2004; Jinno et al., 2005). To validate the specificity of the ) ⁄ ) (10, 168) immunofluorescent reactions, Kv4.2 mice were used. The labelling pattern in control mice (Fig. 1B) was very similar to that obtained in 13.3 rats. The complete lack of labelling in the Kv4.2) ⁄ ) mice (Fig. 1C) demonstrates that the immunolabelling at the light microscopic level is due to specific antibody–antigen interactions. To compare quantita-

(10, 155) tively the Kv4.2 subunit content of distinct somato-dendritic compartments, we carried out immunogold labelling using the SDS-FRL 12.1 method. We systematically investigated gold particle densities in main apical dendrites, oblique dendrites and dendritic spines in the inner, middle and outer thirds of the SR, dendritic shafts and spines in the

(8, 105) SLM in addition to the somata of CA1 PCs. 11.0

High-resolution immunogold localization of the Kv4.2 subunit along the CA1 PCs somato-dendritic axis

(10, 286) Electron microscopic analysis of the replicas revealed many gold particles labelling the Kv4.2 subunit on P-faces of somatic and 11.4 dendritic plasma membranes, consistent with the intracellular location of the epitope (AA 454–469) recognized by the rabbit anti-Kv4.2 antibody. Gold particles were apparently randomly distributed and (9, 343) showed a rather similar distribution pattern in the plasma membranes of

11.9 somata and main apical dendrites of CA1 PCs (Fig. 2B–E). Quanti- fication of the immunogold reactions demonstrated a moderate . In parenthesis, the first number indicates the number of analyzed profiles and the second number denotes the number of counted gold particles.

2 distance-dependent increase in gold particle density along the m

l proximo-distal axis of the main apical dendrites within the SR ⁄ (9, 276) (proximal SR – 7.8 ± 3.8 gold ⁄ lm2; middle – 11.9 ± 3.7 gold ⁄ lm2; 2 7.8 distal – 11.4 ± 2.8 gold ⁄ lm ; n = 5 rats; Table 1), with a slight decrease in dendrites of the SLM (8.9 ± 3.4 gold ⁄ lm2). The density of gold particles labelling the Kv4.2 subunit in the proximal apical dendrite was very similar to that found in the somata (6.9 ± 1.7

(9, 307) gold ⁄ lm2; Fig. 2A and A’). The average relative increase from the proximal to distal dendrites within the SR was 69 ± 50% in ﬁve rats.

Densities of gold particles labelling the Kv4.2 subunit in distinct somato-dendritic compartments of rat CA1 PCs To investigate whether a much higher density of the Kv4.2 subunit 0.7 6.9 in oblique dendrites or dendritic spines could mask the slight increase found in main apical dendrites and therefore result in a uniform #2#3#4 0.6#5 1.7 7.8 (10, 352)Mean 0.2 8.5 (10, 13.4 463) (10, 0.1 5.4 352) (8, 163) 6.5 4.8 17.4 (11, (8, (10, 324) 185) 442) 5.0 12.1 15.6 (9, (9, (11, 240) 300) 367) 4.1 (7, 140) 16.3 11.6 (10, 11.0 (8, 191) (10, 296) 256) 17.8 6.9 11.2 (8, (10, (10, 239) 11.3 284) 124) (8, 238) 15.9 18.3 7.7 (10, (13, (9, 183) 292) 241) 7.6 (8, 18.6 19.7 83) (10, (12, 200) 41) 4.6 (5, 17.3 27) (13, 9.0 32) (8, 7) 8.5 (8, 17.3 64) (15, 50) 7.4 (10, 15.7 116) (12, 22) 8.7 (19, 7.7 (9, 16.9 7.6 78) (7, (10, 10) 96) 14.3 (16, 8.8 3.6 395) (12, (96, 16) 3) 14.2 (10, 19) 9.5 (11, 10.6 20) (11, 12) 9.6 (12, 5.7 15) (8, 98) 6.2 (9, 178) 7.6 (5, 8) 6.5 (13, 16) Table 1. RAT BG Soma#1 1.1 8.2 (10, Prox 371) Ap Dend 9.9 Mid (9, Ap 323) Dend Dist Ap 11.6 Dend (10, 394) Prox 11.4 Ob (10, Dend 328) Mid 15.3 Ob (7, Dend 99) Dist Ob Dend 11.0 (12, Prox 172) Sp 14.2 (10, 175) 13.7 (14, Mid 27) Sp 18.7 (24, 87) 13.5 (12, 39) Dist Sp 9.6 (12, 202 SLM Dend SLM Sp SD 0.7Density values are 1.7 provided (1, in 128) gold 3.8 (1, 87) 3.7 (1, 71) 2.8 (1, 58)staining 5.0 (2, 60) of 4.6 the (2,67) SR 5.4 (2, 86)as observed 6.1 (3, 15) 3.5 (6, 29) at the 3.5 (3, 18) light 3.4 (5, 119) microscopic 5.7 (5, 21) level, we

Table 2. Densities of gold particles labelling the Kv4.2 subunit in axon terminals of rat, Kv4.2+ ⁄ + and Kv4.2) ⁄ ) mice CA1 PCs

Rat BG Mid Ap Dend Axon Mouse BG Mid Ap Dend Axon Mouse BG Mid Ap Dend Axon

#1 0.14 7.0 (10, 349) 1.7 (84, 39) Kv4.2+ ⁄ + #1 0.50 6.3 (10, 157) 0.9 (91, 20) Kv4.2) ⁄ ) #1 0.80 1.2 (10, 30) 2.7 (93, 65) #2 0.95 16.6 (7, 220) 2.8 (66, 62) Kv4.2+ ⁄ + #2 0.50 11.3 (9, 303) 4.0 (67, 64) Kv4.2) ⁄ ) #2 0.48 0.3 (7, 5) 1.8 (137, 66) #3 0.40 9.2 (5, 169) 2.6 (39, 44) Kv4.2+ ⁄ + #3 0.52 8.1 (7, 98) 3.0 (169, 110) Kv4.2) ⁄ ) #3 0.48 1.9 (8, 34) 2.1 (98, 50) Mean 0.50 11.0 (7, 246) 2.4 (63, 48) 0.50 8.6 (9, 186) 2.6 (109, 65) 0.59 1.2 (8, 23) 2.2 (109, 60) SD 0.42 5.1 (3, 93) 0.6 (23, 12) 0.01 2.6 (2, 106) 1.6 (53, 45) 0.19 0.8 (2, 16) 0.4 (24, 9)

Density values are provided in gold ⁄ lm2. In parentheses, the ﬁrst number indicates the number of analysed proﬁles and the second number denotes the number of counted gold particles.

Table 3. The immunoreaction is speciﬁc in all somato-dendritic subcellular compartments of CA1 PCs

Mouse BG Soma Ap Dend Ob Dend SLM Dend Spine

Kv4.2) ⁄ ) #1 0.37 0.42 (8, 13) 0.28 (26, 22) 0.47 (31, 11) 0.43 (9, 7) 0.18 (38, 1) Kv4.2) ⁄ ) #2 0.41 0.54 (10, 27) 0.59 (21, 39) 0.45 (22, 10) 0.70 (11, 11) 1.22 (35, 5) Kv4.2) ⁄ ) #3 0.29 1.31 (9, 54) 2.44 (25, 167) 3.29 (28, 82) 2.07 (12, 40) 3.34 (39, 16) Mean 0.36 0.76 (9, 31) 1.10 (24, 76) 1.40 (27, 34) 1.07 (11, 19) 1.58 (37, 7) SD 0.06 0.48 (1, 21) 1.17 (3, 79) 1.63 (5, 41) 0.88 (2, 18) 1.61 (2, 8)

Density values are provided in gold ⁄ lm2. In parentheses, the ﬁrst number indicates the number of analysed proﬁles and the second number denotes the number of counted gold particles.

ABC

Fig. 1. Distribution of the Kv4.2 subunit immunoreactivity in the hippocampus and the specificity of immunoreactions. (A) The immunfluorescent reaction shows strong, homogenous neuropil labelling for the Kv4.2 subunit in strata oriens (so) and radiatum (sr) of rat CA1 region, with a slight reduction of the immunolabelling in stratum lacunosum-moleculare (slm). (B) A very similar labelling pattern was observed in the mouse hippocampus. (C) The labelling was absent in the CA1 area of Kv4.2) ⁄ ) mice, demonstrating the specificity of the immunoreaction. sp, stratum pyramidale. All images are single confocal sections. Scale bars: 50 lm (A–C). quantified gold particle densities in oblique dendrites and dendritic somatic membranes, are above the non-specific labelling (background spines in the above-mentioned three subdivisions of the SR (Fig. 3). – 0.7 ± 0.7 gold ⁄ lm2; anova with Dunnett’s post hoc test, Gold particle densities in oblique dendrites (proximal SR – P < 0.05). Despite the increasing–decreasing tendency in the density 11.0 ± 5.0 gold ⁄ lm2; middle – 12.1 ± 4.6 gold ⁄ lm2; distal – of gold particles along the dendritic regions, statistical comparisons 13.3 ± 5.4 gold ⁄ lm2; n = 5 rats) and in dendritic spines (proximal revealed no significant difference in the densities among these SR – 11.0 ± 6.1 gold ⁄ lm2; middle – 15.1 ± 3.5 gold ⁄ lm2; distal – dendritic compartments (P = 0.08, anova). Previous studies sug- 13.6 ± 3.5 gold ⁄ lm2; SLM – 11.2 ± 5.7 gold ⁄ lm2; n = 5 rats; gested that A-type K+ channels potentially located around dendritic Table 1) showed an almost identical increasing–decreasing pattern branch points have a critical role in gating the propagation of dendritic to that found for the main apical trunks (Fig. 4). In each subregion of spikes towards the soma or into other dendrites (Cai et al., 2004; the SR, the average densities in oblique dendrites and spines were only Losonczy et al., 2008). Therefore, we specifically investigated the 26 ± 16% higher than those found in the main apical trunks, distribution pattern and density of the immunogold particles for the demonstrating the lack of large quantitative differences in the densities Kv4.2 subunit on plasma membranes of dendritic branch points, but of Kv4.2 subunit among these distinct dendritic compartments. The could not detect any clustering or increase in particle density (Fig. 2F densities of gold particles in all examined compartments, but the and F’).

A B C

A′ B′ C′

D E F

D′ E′ F′

Fig. 2. High-resolution immunogold localization of the Kv4.2 subunit in somata and apical dendrites of CA1 PCs. (A–E) Low-magnification images of P-faces of a PC soma (A) and spiny apical dendrites from the proximal (B), middle (C) and distal (D) stratum radiatum and stratum lacunosum-moleculare (E) show a rather uniform immunogold labelling pattern. (A¢–E¢) High-magnification images of the boxed region in (A–E). (F) Immunogold particles for the Kv4.2 subunit are homogenously distributed around a branch point on a PC dendrite. (F¢) High-magnification view of the boxed region in (F). dist. sr, distal stratum radiatum; mid. sr, middle stratum radiatum; prox. sr, proximal stratum radiatum; slm, stratum lacunosum-moleculare; sp, stratum pyramidale. Scale bars: 1 lm (A); 500 nm (B–F); 100 nm (A¢–F¢).

Specificity of the Kv4.2 subunit immunogold labelling in CA1 unpaired Student’s t-test) to that obtained in five rats, as assessed from PCs using SDS-FRL the gold particle densities in the main apical dendrites in the middle of To validate the specificity of immunogold labelling on SDS-FRL, we the SR (8.6 ± 2.6 gold ⁄ lm2, n = 3 control mice; Fig. 5A; Table 2), repeated our reactions in control and Kv4.2) ⁄ ) mice (Fig. 5A and B). and showed a similar labelling pattern within the SR and SLM to that In control mice, the strength of the reaction was similar (P = 0.22, found in rats. In contrast, in Kv4.2) ⁄ ) mice, the mean density of gold

A B C

Fig. 3. Distribution of immunogold particles for the Kv4.2 subunit in oblique dendrites and spines of CA1 PCs in the stratum radiatum. (A) Immunogold particles are homogenously distributed along the P-face of an oblique dendrite (obl.d.) in the proximal stratum radiatum (prox. sr). (B) Slightly higher immunogold particle density for the Kv4.2 subunit can be seen on an oblique dendrite from the middle stratum radiatum (mid. sr). Note that a spine (sp) with high density of immunogold particles emerges from the dendrite and faces an axon terminal (at) E-face. (C) A spiny oblique dendrite in the distal stratum radiatum (dist. sr) contains a large number of immunogold particles for the Kv4.2 subunit. Scale bars: 250 nm (A–C).

supraoptic nucleus (Alonso & Widmer, 1997), developing cerebellum (Shibasaki et al., 2004), visual cortex (Burkhalter et al., 2006) and subiculum (Jinno et al., 2005). The SDS-FRL technique allows the visualization of neurotransmitter receptors and their associated proteins in both GABAergic (Kasugai et al., 2010; Lorincz & Nusser, 2010) and glutamatergic synapses (Kulik et al., 2006; Masugi-Tokita et al., 2007). We colocalized the Kv4.2 subunit with PSD-95, which clearly marks the PSD of excitatory synapses on the P-face of the replica (Kulik et al., 2006) and with the GABAA receptor b3 subunit (Kasugai et al., 2010; Lorincz & Nusser, 2010), which labels inhibitory synapses also on the P-face. We could not find any evidence for the enrichment of Kv4.2 subunits in GABAergic or glutamatergic synapses, only a few scattered gold particles were found around the periphery of some excitatory synapses (Fig. 6A). Immu- nogold particles labelling the Kv4.2 subunits infrequently clustered on somatic and dendritic membranes, but they never colocalized with the GABA receptor b3 subunit (Fig. 6B). These results provide evidence Fig. 4. Densities of Kv4.2 immunogold particles in different somato-dendritic A compartments of rat CA1 PCs. Bar graphs illustrate the background subtracted for the lack of synaptic accumulation of the Kv4.2 subunit in CA1 densities (mean ± SD, in gold ⁄ lm2) of immunogold particles in the somato- PCs. dendritic compartments. The bars are colour coded to different subcellular compartments as indicated in the schematic drawing of a CA1 PC. Note the moderate increase in the density along the proximo-distal axis of PCs within the A low density of immunogold particles for the Kv4.2 subunit stratum radiatum and the subsequent decrease in the stratum lacunosum- moleculare. was found in SNAP-25-containing axon terminals, but it persisted in Kv4.2) ⁄ ) mice Finally, we noticed that gold particles for the Kv4.2 subunit were not particles on P-face structures (soma – 0.76 ± 0.48 gold ⁄ lm2; apical 2 only confined to somato-dendritic plasma membranes, but were dendrite – 1.10 ± 1.17 gold ⁄ lm ; oblique dendrite – present in presumed presynaptic boutons at a low density. To 1.40 ± 1.63 gold ⁄ lm2; SLM – 1.07 ± 0.88 gold ⁄ lm2; dendritic spine 2 unequivocally identify these weakly labelled structures, we performed – 1.58 ± 1.61 gold ⁄ lm ; n = 3 mice; Fig. 5B; Table 3) was not double-labelling experiments for Kv4.2 and SNAP-25, a member of significantly different from that obtained in the E-face structures the SNARE complex present exclusively in axons (Hagiwara et al., around them (0.36 ± 0.06 gold ⁄ lm2; n =3;P = 0.08, anova). The ) ⁄ ) 2005). The strength of the Kv4.2 labelling in these double-labelling low insignificant labelling in Kv4.2 tissue was consistent in all reactions in apical dendrites from the middle SR analysed somato-dendritic compartments throughout the depth of the (10.9 ± 5.1 gold ⁄ lm2; n = 3 rats) was very similar to that found in SR and SLM (Table 3), demonstrating the specificity of immunogold single-labelling reactions (P = 0.77; unpaired Student’s t-test; Ta- labelling in all somato-dendritic compartments. ble 1). The SNAP-25-positive structures contained on average 2.4 ± 0.6 gold particles in each lm2 (n = 3 rats; Fig. 5C; Table 2). Although this density is only a quarter of that found on the somato- The Kv4.2 subunit is excluded from the postsynaptic membrane dendritic compartments, it is still significantly (P = 0.008, paired specializations Student’s t-test) higher than the background labelling Next, we asked whether the Kv4.2 subunit is concentrated in (0.5 ± 0.4 gold ⁄ lm2). The presence of significant immunolabelling GABAergic or glutamatergic synapses, as suggested to occur in the in axon terminals is surprising, because the Kv4.2 subunit is

A B

C D E

Fig. 5. Specificity test for Kv4.2 subunit labelling on the axo-somato-dendritic surface of CA1 PCs using SDS-FRL. (A) Electron micrograph illustrating the P-face of a spiny CA1 PC dendrite of wild-type mouse immunolabelled for the Kv4.2 subunit. Note that the density and distribution pattern of the immunogold labelling for the Kv4.2 subunit in mouse is similar to that seen in rat. (B) The lack of immunogold labelling for the Kv4.2 subunit in a spiny dendrite of a Kv4.2) ⁄ ) mouse demonstrates the specificity of the labelling using SDS-FRL. (C) Axon terminals in rat identified by immunolabelling for SNAP-25 (15-nm gold) contain a low number of immunogold particles (arrows) for the Kv4.2 subunit (10-nm gold). (D and E) Immunogold particles for the Kv4.2 subunit were present in both Kv4.2+ ⁄ + (D) and Kv4.2) ⁄ ) mice (E) at approximately the same density. dendr, dendrite; sp, spine. Scale bars: 250 nm (A–E).

A B

Fig. 6. The Kv4.2 subunit is excluded from postsynaptic membrane specializations. (A) Electron micrograph illustrating an excitatory synapse, revealed by the accumulation of the postsynaptic density marker PSD-95 (15-nm gold). Arrowheads point to gold particles labelling the Kv4.2 subunit (10-nm gold) around the PSD. (B) High-magnification image of an inhibitory synapse identified by the enrichment of gold particles (5-nm gold) labelling the c-aminobutyric acid (GABA)A receptor b3 subunit. Note the clustering of gold particles labelling the Kv4.2 subunit (10-nm gold) in the extrasynaptic membrane, but not within the inhibitory synapse. Scale bars: 100 nm (A); 50 nm (B). conceived as a somato-dendritic ion channel. To confirm the control mice (2.6 ± 1.6 gold ⁄ lm2; n = 3 mice; Fig. 5D; Table 2) specificity of axon terminal immunolabelling, we repeated these was very similar (P = 0.82, unpaired Student’s t-test) to that found in experiments in control and Kv4.2) ⁄ ) mice. The labelling intensity in rats, but surprisingly the density of gold particles labelling the Kv4.2

ª 2011 The Authors. European Journal of Neuroscience ª 2011 Federation of European Neuroscience Societies and Blackwell Publishing Ltd European Journal of Neuroscience, 35, 66–75 Somato-dendritic distribution of Kv4.2 on CA1 PCs 73 subunit was almost identical in the axon terminals of control and K-channel-interacting proteins and dipeptidyl peptidase-like type II ) ) Kv4.2 ⁄ mice (axon terminals – 2.2 ± 0.4 gold ⁄ lm2; n = 3 mice; proteins, DPP6 and DPP10 (reviewed by Vacher et al., 2008). It P = 0.69, unpaired Student’s t-test; Fig. 5E; Table 2). This value was remains to be seen whether the interaction with these or other auxiliary significantly (P = 0.004, paired Student’s t-test) higher than the proteins could completely silence Kv4.2 channels, and whether such background labelling on the surrounding E-face structures proteins have an inverse distance-dependent distribution along the (0.6 ± 0.2 gold ⁄ lm2). Thus, our results reveal that the immunogold proximo-distal axes of PC dendrites. labelling in the somato-dendritic compartments is due to specific Our high-resolution immunolocalization experiments revealed a antibody–Kv4.2 subunit interactions, whereas the same antibody novel, distance- and subcellular compartment-specific distribution under identical experimental conditions provides a weak, non-specific pattern of an ion channel on the surface of CA1 PCs. Previously, our labelling in axon terminals. laboratory has examined the cell surface distribution of the hyperpo- larization-activated mixed cation channel (HCN) subunit 1 in hippocampal, subicular and neocortical layer 5 PCs, and found an exclusive Discussion somato-dendritic distribution. It was reported that the density of In the present study, we have applied a highly sensitive, high- HCN1 increases as a function of distance along the proximo-distal resolution quantitative electron microscopic immunogold technique to axis of the PC dendrite, with a steep, supralinear increase in the SLM investigate the densities of the Kv4.2 subunit in 13 distinct axo- (Lorincz et al., 2002). Although the density of Ih, mediated by somato-dendritic compartments of CA1 PCs. Our results revealed a HCN1 ⁄ HCN2, and IA, mediated by the Kv4.2 subunit, is very similar, previously unseen distribution pattern on the surface of PCs, where the the subcellular distributions of the underlying channel subunits are density of Kv4.2 subunit shows a slight increase from proximal to remarkably different. Instead of having a monotonous increase, the distal dendrites in the SR, then a decrease in the SLM. The density of the Kv4.2 subunit first increases slightly within the SR, then approximately 70% increase in the density within the SR is much decreases in the SLM. Both of these distribution patterns are very + less than the reported increase in IA along the main apical trunk. different from that of the Nav1.6 subunit of the voltage-gated Na Although the density of Kv4.2 subunit is higher in small-diameter channels, which shows a distance-dependent decrease from proximal oblique dendrites and dendritic spines than that in the corresponding to distal dendrites of CA1 PCs (Lorincz & Nusser, 2010). Although main apical trunk, this moderate, approximately 25%, difference is our laboratory have carried out detailed quantitative electron micro- unlikely to underlie a fundamentally distinct role of IA in main vs scopic analysis of the subcellular distribution of only these three oblique dendrites. voltage-gated ion channel subunits, light-microscopic immunolocal- Our results demonstrating only a moderate distance-dependent izations indicate that a large number of ion channel subunits (e.g. increase in the density of the Kv4.2 subunit from proximal to distal Kv1.1, Kv1.2, Kv2.1) have their unique cell surface distribution dendrites in SR are in disagreement with results obtained with patterns within the same cell type (CA1 PCs). These data, taken dendritic recordings of IA. Hoffman et al. (1997) have demonstrated a together, prompt the hypothesis that each ion channel subunit has its approximately sixfold increase in peak IA within the same dendritic unique subcellular distribution pattern on the surface of a given nerve segments of CA1 PCs. What could be the reason for this discrepancy cell. between IA and Kv4.2 subunit, when dendritic IA is exclusively Our results, showing the specificity of immunogold signal on the mediated by this subunit? A possible explanation could be that Kv4.2 somato-dendritic compartments and the lack of it in axon terminals, subunits are in distinct functional states along the apical dendrites of demonstrate that a given antibody could provide specific and non- PCs; i.e. a large fraction of the channels in the proximal dendrites do specific labelling in distinct subcellular compartments under identical not conduct K+ within a functionally relevant voltage range. We thus experimental conditions. Although it seems like a minor technical have to assume that certain posttranslational modifications or inter- issue, it has far reaching consequences in immunohistochemistry. It is actions with some auxiliary ⁄ associated proteins might regulate the also important to note that immunosignal obtained with light function of these ion channels in a dramatic manner. The most widely microscopic fluorescent techniques completely disappeared in studied posttranslational modulation of Kv4.2 channels is phosphor- Kv4.2) ⁄ ) mice, demonstrating that all immunosignals in these light ylation. It has been shown that Kv4.2 channels and IA in hippocampal microscopic reactions were due to specific antibody–Kv4.2 protein CA1 PC dendrites are regulated by protein kinase C (PKC), protein interactions. If we had taken these light microscopic control exper- kinase A (PKA), CaMKII and mitogen-activated protein kinase iments as evidence for the specificity of our immunogold signal in (Hoffman & Johnston, 1998; Yuan et al., 2002). Phosphorylation by SDS-FRL, we would have erroneously concluded the presynaptic PKC and PKA results in a 15-mV depolarizing shift in the activation presence of the Kv4.2 subunit. These results provide further evidence curve of IA in CA1 PC dendrites (Hoffman & Johnston, 1998). As a that the concept of ‘antibody specificity’ is erroneous. If the specificity result, the same depolarization activates less IA following PKA ⁄ PKC was the ‘property’ of the primary antibody, then either all immuno- phosphorylation of the channels. Therefore, a distance-dependent signal under all experimental conditions should be specific or should decrease in the phosphorylation state of the Kv4.2 subunit along the be non-specific. This is clearly not the case. Watanabe et al. (1998) proximo-distal dendrites of CA1 PCs could reconcile the differences provided a compelling demonstration of this. When an antibody in the activation curve of IA in distal and proximal dendrites (Hoffman against the NR2A subunit was used under conventional conditions for et al., 1997), but could not explain the differences in the peak IA light microscopic localization, the somatic and proximal apical density increase and that of Kv4.2 density in main apical trunks within dendritic cytoplasm of CA3 PCs became strongly labelled, a pattern the SR. It should also be noted that phosphorylation by PKA also that was identical in NR2A) ⁄ ) mice. However, when the tissue was causes internalization of Kv4.2 channels from spine plasma mem- treated with pepsin before the immunoreactions, the staining pattern branes of PCs (Kim et al., 2007; Hammond et al., 2008). Such changed to a punctate neuropil labelling, and the strong cytoplasmic regulation, however, results in changes in surface density of the labelling disappeared. This labelling pattern following antigen channels, which is readily detectable with SDS-FRL. Alternatively, retrieval was completely absent in NR2A) ⁄ ) mice, demonstrating the function of Kv4.2 channels might be modulated by protein–protein that the same primary antibody could provide specific and non-specific interactions. Kv4 channels have been demonstrated to interact with immunolabelling under different experimental conditions. Similar to

ª 2011 The Authors. European Journal of Neuroscience ª 2011 Federation of European Neuroscience Societies and Blackwell Publishing Ltd European Journal of Neuroscience, 35, 66–75 74 K. Kerti et al. this, the primary antibody used in the present study provided a strong Gutman, G.A., Chandy, K.G., Adelman, J.P., Aiyar, J., Bayliss, D.A., cytoplasmic cell body staining for Burkhalter et al. (2006), which Clapham, D.E., Covarriubias, M., Desir, G.V., Furuichi, K., Ganetzky, B., remained very similar in Kv4.2) ⁄ ) mouse tissue. The same primary Garcia, M.L., Grissmer, S., Jan, L.Y., Karschin, A., Kim, D., Kuperschmidt, S., Kurachi, Y., Lazdunski, M., Lesage, F., Lester, H.A., McKinnon, D., antibody under our experimental conditions provided an exclusive Nichols, C.G., O’Kelly, I., Robbins, J., Robertson, G.A., Rudy, B., neuropil labelling in the hippocampus and neocortex, which com- Sanguinetti, M., Seino, S., Stuehmer, W., Tamkun, M.M., Vandenberg, pletely disappeared in Kv4.2) ⁄ ) mouse tissue, providing further C.A., Wei, A., Wulff, H. & Wymore, R.S. (2003) International Union of evidence that the experimental conditions are crucial when the Pharmacology. XLI. Compendium of voltage-gated ion channels: potassium channels. Pharmacol. Rev., 55, 583–586. outcome of an immunoreaction is concerned. Furthermore, our SDS- Hagiwara, A., Fukazawa, Y., Deguchi-Tawarada, M., Ohtsuka, T. & Shigem- FRL experiments demonstrated that even under identical reaction oto, R. (2005) Differential distribution of release-related proteins in the conditions, a primary antibody could provide both specific and non- hippocampal CA3 area as revealed by freeze-fracture replica labelling. specific labelling in different subcellular compartments. This is not J. Comp. Neurol., 489, 195–216. surprising, because the cross-reacting molecule ⁄ protein could have a Hammond, R.S., Lin, L., Sidorov, M.S., Wikenheiser, A.M. & Hoffman, D.A. (2008) Protein kinase A mediates activity-dependent Kv4.2 channel subcellular compartment-specific distribution pattern within the trafficking. J. Neurosci., 28, 7513–7519. examined cell. Our results provide an alarming example, and Hille, B. (2001). Ionic Channels of Excitable Membranes. Sinauer Associates emphasize that the specificity of immunosignal must be proven under Inc., Sunderland, USA. + each experimental condition in each brain region, cell type and even Hoffman, D.A. & Johnston, D. (1998) Downregulation of transient K channels in dendrites of hippocampal CA1 pyramidal neurons by activation of PKA subcellular compartment (Lorincz & Nusser, 2008). and PKC. J. Neurosci., 18, 3521–3528. Hoffman, D.A., Magee, J.C., Colbert, C.M. & Johnston, D. (1997) K+ channel regulation of signal propagation in dendrites of hippocampal pyramidal Acknowledgements neurons. Nature, 387, 869–875. Jinno, S., Jeromin, A. & Kosaka, T. (2005) Postsynaptic and extrasynaptic A.L. is the recipient of a Jańos Bolyai Scholarship from the Hungarian localization of Kv4.2 channels in the mouse hippocampal region, with special Academy of Sciences. Z.N. is the recipient of a Project (090197 ⁄ Z ⁄ 09 ⁄ Z) and reference to targeted clustering at gabaergic synapses. Neuroscience, 134, an Equipment (083484 ⁄ Z ⁄ 07 ⁄ Z) Grant from the Wellcome Trust, a Hungarian 483–494. National Office for Research and Technology-French National Research Kasugai, Y., Swinny, J.D., Roberts, J.D., Dalezios, Y., Fukazawa, Y., Sieghart, Agency Te´T Fund (NKTH-Neurogen) and a European Young Investigator W., Shigemoto, R. & Somogyi, P. (2010) Quantitative localisation of Award (http://www.esf.org/euryi). The financial support from these Founda- synaptic and extrasynaptic GABAA receptor subunits on hippocampal tions is gratefully acknowledged. We would like to thank E´ va Dobai and Do´ra pyramidal cells by freeze-fracture replica immunolabelling. Eur. J. Neuro- Rońasze´ki for their excellent technical assistance. We would like to thank Prof. sci., 32, 1868–1888. ) ⁄ ) Daniel Johnston and Dr Darrin Brager for kindly providing the Kv4.2 mice. Kim, J. & Hoffman, D.A. (2008) Potassium channels: newly found players in synaptic plasticity. Neuroscientist, 14, 276–286. Kim, J., Jung, S.C., Clemens, A.M., Petralia, R.S. & Hoffman, D.A. (2007) Abbreviations Regulation of dendritic excitability by activity-dependent trafficking of the A-type K+ channel subunit Kv4.2 in hippocampal neurons. Neuron, 54, 933– BSA, bovine serum albumin; GABA, c-aminobutyric acid; HCN, hyperpolar- 947. + ization-activated mixed cation channel; IA, A-type K current; NGS, normal Kulik, A., Vida, I., Fukazawa, Y., Guetg, N., Kasugai, Y., Marker, C.L., goat serum; PB, phosphate buffer; PC, pyramidal cell; PKA, protein kinase A; Rigato, F., Bettler, B., Wickman, K., Frotscher, M. & Shigemoto, R. (2006) PKC, protein kinase C; PSD, postsynaptic density; SDS-FRL, sodium dodecyl Compartment-dependent colocalization of Kir3.2-containing K+ channels sulphate-digested freeze-fracture replica-labelling; SLM, stratum lacunosum- and GABAB receptors in hippocampal pyramidal cells. J. Neurosci., 26, moleculare; SP, stratum pyramidale; SR, stratum radiatum; TBS, Tris-buffered 4289–4297. saline. Lorincz, A. & Nusser, Z. (2008) Specificity of immunoreactions: the importance of testing specificity in each method. J. Neurosci., 28, 9083– 9086. Lorincz, A. & Nusser, Z. (2010) Molecular identity of dendritic voltage-gated References sodium channels. Science (New York, N.Y), 238, 906–909. Alonso, G. & Widmer, H. (1997) Clustering of KV4.2 potassium channels in Lorincz, A., Notomi, T., Tamas, G., Shigemoto, R. & Nusser, Z. (2002) postsynaptic membrane of rat supraoptic neurons: an ultrastructural study. Polarized and compartment-dependent distribution of HCN1 in pyramidal Neuroscience, 77, 617–621. cell dendrites. Nat. Neurosci., 5, 1185–1193. Birnbaum, S.G., Varga, A.W., Yuan, L.L., Anderson, A.E., Sweatt, J.D. & Losonczy, A., Makara, J.K. & Magee, J.C. (2008) Compartmentalized Schrader, L.A. (2004) Structure and function of Kv4-family transient dendritic plasticity and input feature storage in neurons. Nature, 452, 436– potassium channels. Physiol. Rev., 84, 803–833. 441. Burkhalter, A., Gonchar, Y., Mellor, R.L. & Nerbonne, J.M. (2006) Differential Makara, J.K., Losonczy, A., Wen, Q. & Magee, J.C. (2009) Experience- expression of IA channel subunits Kv4.2 and Kv4.3 in mouse visual cortical dependent compartmentalized dendritic plasticity in rat hippocampal CA1 neurons and synapses. J. Neurosci., 26, 12274–12282. pyramidal neurons. Nat. Neurosci., 12, 1485–1487. Cai, X., Liang, C.W., Muralidharan, S., Kao, J.P., Tang, C.M. & Thompson, Maletic-Savatic, M., Lenn, N.J. & Trimmer, J.S. (1995) Differential spatio- S.M. (2004) Unique roles of SK and Kv4.2 potassium channels in dendritic temporal expression of K+ channel polypeptides in rat hippocampal neurons integration. Neuron, 44, 351–364. developing in situ and in vitro. J. Neurosci., 15, 3840–3851. Cash, S. & Yuste, R. (1999) Linear summation of excitatory inputs by CA1 Margulis, M. & Tang, C.M. (1998) Temporal integration can readily switch pyramidal neurons. Neuron, 22, 383–394. between sublinear and supralinear summation. J. Neurophysiol., 79, 2809– Chen, X., Yuan, L.L., Zhao, C., Birnbaum, S.G., Frick, A., Jung, W.E., 2813. Schwarz, T.L., Sweatt, J.D. & Johnston, D. (2006) Deletion of Kv4.2 gene Masugi-Tokita, M. & Shigemoto, R. (2007) High-resolution quantitative eliminates dendritic A-type K+ current and enhances induction of long-term visualization of glutamate and GABA receptors at central synapses. Curr. potentiation in hippocampal CA1 pyramidal neurons. J. Neurosci., 26, Opin. Neurobiol., 17, 387–393. 12143–12151. Masugi-Tokita, M., Tarusawa, E., Watanabe, M., Molnar, E., Fujimoto, K. & Frick, A., Magee, J. & Johnston, D. (2004) LTP is accompanied by an Shigemoto, R. (2007) Number and density of AMPA receptors in individual enhanced local excitability of pyramidal neuron dendrites. Nat. Neurosci., 7, synapses in the rat cerebellum as revealed by SDS-digested freeze-fracture 126–135. replica labelling. J. Neurosci., 27, 2135–2144. Fujimoto, K. (1995) Freeze-fracture replica electron microscopy combined with Migliore, M., Hoffman, D.A., Magee, J.C. & Johnston, D. (1999) Role of an SDS digestion for cytochemical labelling of integral membrane proteins. A-type K+ conductance in the back-propagation of action potentials in Application to the immunogold labelling of intercellular junctional com- the dendrites of hippocampal pyramidal neurons. J. Comput. Neurosci., 7,5– plexes. J. Cell Sci., 108, 3443–3449. 15.

Rash, J.E., Yasumura, T., Hudson, C.S., Agre, P. & Nielsen, S. (1998) Direct Vacher, H., Mohapatra, D.P. & Trimmer, J.S. (2008) Localization and targeting immunogold labelling of aquaporin-4 in square arrays of astrocyte and of voltage-dependent ion channels in mammalian central neurons. Physiol. ependymocyte plasma membranes in rat brain and spinal cord. Proc. Natl. Rev., 88, 1407–1447. Acad. Sci. USA, 95, 11981–11986. Varga, A.W., Anderson, A.E., Adams, J.P., Vogel, H. & Sweatt, J.D. (2000) Rhodes, K.J., Carroll, K.I., Sung, M.A., Doliveira, L.C., Monaghan, M.M., Input-specific immunolocalization of differentially phosphorylated Kv4.2 in Burke, S.L., Strassle, B.W., Buchwalder, L., Menegola, M., Cao, J., An, the mouse brain. Learn. Mem., 7, 321–332. W.F. & Trimmer, J.S. (2004) KChIPs and Kv4 alpha subunits as integral Watanabe, M., Fukaya, M., Sakimura, K., Manabe, T., Mishina, M. & Inoue, components of A-type potassium channels in mammalian brain. J. Neurosci., Y. (1998) Selective scarcity of NMDA receptor channel subunits in the 24, 7903–7915. stratum lucidum (mossy fibre-recipient layer) of the mouse hippocampal Serodio, P. & Rudy, B. (1998) Differential expression of Kv4 K+ channel CA3 subfield. Eur. J. Neurosci., 10, 478–487. subunits mediating subthreshold transient K+ (A-type) currents in rat brain. Watanabe, S., Hoffman, D.A., Migliore, M. & Johnston, D. (2002) Dendritic K+ J. Neurophysiol., 79, 1081–1091. channels contribute to spike-timing dependent long-term potentiation in Shibasaki, K., Nakahira, K., Trimmer, J.S., Shibata, R., Akita, M., Watanabe, hippocampal pyramidal neurons. Proc. Natl. Acad. Sci. USA, 99, 8366–8371. S. & Ikenaka, K. (2004) Mossy fibre contact triggers the targeting of Kv4.2 Yuan, L.L., Adams, J.P., Swank, M., Sweatt, J.D. & Johnston, D. (2002) potassium channels to dendrites and synapses in developing cerebellar Protein kinase modulation of dendritic K+ channels in hippocampus involves granule neurons. J. Neurochem., 89, 897–907. a mitogen-activated protein kinase pathway. J. Neurosci., 22, 4860–4868.

ª 2011 The Authors. European Journal of Neuroscience ª 2011 Federation of European Neuroscience Societies and Blackwell Publishing Ltd European Journal of Neuroscience, 35, 66–75